Restore Oracle database by manual Backup

Restore the Oracle database by manual backup:

Although the Rman tool is already available

But restoring Oracle manually will give you a deeper understanding of Oracle Database

Database consistent boot conditions:
Data file SCN, control file Scn,redo SCN consistent


Control file Record:
SCN to which data files should arrive
Current Redo
Physical structure information of the data
Archive information

　　

Prerequisites:

Archive Log On

Data files are backed up

Control file has backup

Backup data:

Data file Backup:

Data file into backup mode:

Select ' Alter tablespace ' | | tablespace_name| | ' begin backup; ' from dba_tablespaces where tablespace_name does like '%temp% ';

Confirm access to Backup status:

　　　　

The operating system performs the physical copy command:

CP *.dbf/home/oracle/sg_backup_dbf/

Copy complete, exit Backup mode:

　　　　

Data file backup complete!!!

Backup complete switch several times archive (multiple switching)

alter system switch logfile;

Control file Backup:

Backup into binary form:

Sql>alter database backup Controlfile to '/home/oracle/backup_ctl/prod4_ctl_20170703 ';

To back up a logical file:

sql> ALTER DATABASE backup Controlfile to trace as '/home/oracle/backup_ctl/prod4.ctl ';

　　　　　

One, table space data file deletion recovery:

Restart data error:

　　

Follow the prompts to view the data file no longer exists (deleted)

Restore by copying a previously backed up data file:

cp/home/oracle/sg_backup_dbf/*.

View SCN number: (Inconsistent status)

　　

Data recovery by executing recover database ()

　　

Start the database:

sql> ALTER DATABASE open;

#恢复完成!

Second, the control file is deleted:

To restart the database:

　　

Recover from old version data file:

cp/home/oracle/backup_ctl/prod4_ctl_20170703./control01.ctl

cp/home/oracle/backup_ctl/prod4_ctl_20170703./control02.ctl

Restart again:

　　

Perform recovery:

sql> Recover database using Backup controlfile until cancel;

#auto mode

Error:

(The old control file does not know the situation of redo)

　　　　　　

Perform the recovery again:

sql> Recover database using Backup controlfile until cancel;

#Specify log: {<ret>=suggested | AUTO | CANCEL}

#/u01/app/oracle/oradata/prod4/prod4/redo01.log #使用指定的redo文件恢复 (enter this line manually)

Boot:

sql> altr database open resetlogs;

Recovery complete!

Control file corruption requires a full standby after recovery

Use ALTR database open resetlogs; When booting, Redolog is reinitialized,

Archive log, backup of previous data files is all invalid

Delete Useless archive log

Third, the current redo damage

Restarting the database does not boot

Recovery:

Delete a data file

Recovering a data file using an old backup

The recovery is similar to the previous mentioned:

Restore command:

Recover database using Backup controlfile until cancel;

#auto
Recover database using Backup controlfile until cancel;
#cancel

Delete a control file

Close the database

Use Backup logic to control File Recovery control files (Find a section similar to the following, directly executed)

(A data file is required to recover the control file, but it is not possible to use the current data file, which is why the data file was deleted before it was recovered)

　　　　　　

Perform power on:

ALTER DATABASE open resetlogs;

Fully prepared
Delete Useless archive log

Summarize:

The principle of consistent boot

SCN number

Steps to boot

Required files for boot

　　

Restore Oracle database by manual Backup